Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

EU Grabs ARM for First ExaFLOP Supercomputer

hpcwire.com

Post reply on HN
Print view View original

71–80 of 115 posts

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#71

Post by cmdrk »

londons_explore wrote: post #70
jefft255 wrote: post #13

Earlier quoted context omitted.

Yes, CPUs are still the main workhorse for many scientific workloads. Sometimes just because the code hasn’t been ported, sometimes because it’s just not something that a GPU can do well.

> just because the code hasn’t been ported, Seems stupid to use millions of dollars of supercomputer time just because you can't be bothered to get a few phd students to spend a few months rewriting in CUDA...

sometimes the code is deeply complex stuff that has accumulated for over 30 years. to _just_ rewrite it in CUDA can be a massive undertaking that could easily produce subtly incorrect results that end up in papers could propagate far into the future by way of citations etc

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#72

Post by throwaway10965 »

cmdrk wrote: post #71
londons_explore wrote: post #70

Earlier quoted context omitted.

> just because the code hasn’t been ported, Seems stupid to use millions of dollars of supercomputer time just because you can't be bothered to get a few phd students to spend a few months rewriting in CUDA...

sometimes the code is deeply complex stuff that has accumulated for over 30 years. to _just_ rewrite it in CUDA can be a massive undertaking that could easily produce subtly incorrect results that end up in papers could propagate far into the future by way of citations etc

Sounds like a great job for LLMs. Are there any public repositories of this code? I want to try.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#73

Post by londons_explore »

cmdrk wrote: post #71
londons_explore wrote: post #70

Earlier quoted context omitted.

> just because the code hasn’t been ported, Seems stupid to use millions of dollars of supercomputer time just because you can't be bothered to get a few phd students to spend a few months rewriting in CUDA...

sometimes the code is deeply complex stuff that has accumulated for over 30 years. to _just_ rewrite it in CUDA can be a massive undertaking that could easily produce subtly incorrect results that end up in papers could propagate far into the future by way of citations etc

All the more reason to rewrite it... You don't want some mistake in 30 year old COBOL code to be making your 2023 experiment to have wrong results.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#74

Post by bee_rider »

devnullbrain wrote: post #67
bee_rider wrote: post #52

Earlier quoted context omitted.

I thought SoftBank bought them, making them a Japanese company? In any case, SiPearl seems to be the one designing the actual chip, they are French.

>I thought SoftBank bought them, making them a Japanese company? What a weird metric for determining the nationality of a company. Intel are publicly traded: are they stateless?

I should have been more direct.

The chip is being designed by a French company, they can license the IP from outside the EU while still building up the EU domestic chip building capabilities. They’ve just outsourced one (big) piece of the puzzle.

Calling ARM a Japanese company was just to highlight the international nature of these sorts of projects.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#75

Post by wmf »

chasil wrote: post #62
datameta wrote: post #22

A bit sad that while POWER9 processors were used in then-SoTA supercomputers, as far as I can tell POWER10 (which I worked on more) is not being used for scientific/industrial HPC.

I understand that POWER9 was much more open than its successor. Is that a factor?

No, I think IBM just gave up on HPC.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#76

Post by gmueckl »

londons_explore wrote: post #73
cmdrk wrote: post #71

Earlier quoted context omitted.

sometimes the code is deeply complex stuff that has accumulated for over 30 years. to _just_ rewrite it in CUDA can be a massive undertaking that could easily produce subtly incorrect results that end up in papers could propagate far into the future by way of citations etc

All the more reason to rewrite it... You don't want some mistake in 30 year old COBOL code to be making your 2023 experiment to have wrong results.

That's the complete opposite of what is actually the case: some of that really old code in these programs is battle-tested and verified. Any rewrite of such parts would just destroy that work for no good reason.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#77

Post by stonogo »

wmf wrote: post #75
chasil wrote: post #62

Earlier quoted context omitted.

I understand that POWER9 was much more open than its successor. Is that a factor?

No, I think IBM just gave up on HPC.

Can confirm; supercomputers don't slot neatly enough into quartly EPS goals.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#78

Post by mlyle »

londons_explore wrote: post #70
jefft255 wrote: post #13

Earlier quoted context omitted.

Yes, CPUs are still the main workhorse for many scientific workloads. Sometimes just because the code hasn’t been ported, sometimes because it’s just not something that a GPU can do well.

> just because the code hasn’t been ported, Seems stupid to use millions of dollars of supercomputer time just because you can't be bothered to get a few phd students to spend a few months rewriting in CUDA...

A supercomputer might cost $200M and use $6M of electricity per year.

Amortizing the supercomputer over 5 years, a 12 hour job on that supercomputer may cost $63k.

If you want it cheaper, your choices are:

A) run on the supercomputer as-is, and get your answer in 12 hours (+ scheduling time based on priority)

B) run on a cheaper computer for longer-- an already-amortized supercomputer, or non-supercomputing resources (pay calendar time to save cost)

C) try to optimize the code (pay human time and calendar time to save cost) -- how much you benefit depends upon labor cost, performance uplift, and how much calendar time matters.

Not all kinds of problems get much uplift from CUDA, anyways.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#79

Post by otabdeveloper4 »

londons_explore wrote: post #70
jefft255 wrote: post #13

Earlier quoted context omitted.

Yes, CPUs are still the main workhorse for many scientific workloads. Sometimes just because the code hasn’t been ported, sometimes because it’s just not something that a GPU can do well.

> just because the code hasn’t been ported, Seems stupid to use millions of dollars of supercomputer time just because you can't be bothered to get a few phd students to spend a few months rewriting in CUDA...

CUDA is buggy proprietary shit that doesn't work half the time or segfaults with compiler errors.

Basically, unless you have a very specific workload that NVidia has specifically tested, I wouldn't bother with it.

Re: EU Grabs ARM for First ExaFLOP Supercomputer

#80

Post by mlyle »

londons_explore wrote: post #73
cmdrk wrote: post #71

Earlier quoted context omitted.

sometimes the code is deeply complex stuff that has accumulated for over 30 years. to _just_ rewrite it in CUDA can be a massive undertaking that could easily produce subtly incorrect results that end up in papers could propagate far into the future by way of citations etc

All the more reason to rewrite it... You don't want some mistake in 30 year old COBOL code to be making your 2023 experiment to have wrong results.

The whole point is in these older numerical codes is that they're proven and there's a long history of results to compare against.
Post reply on HN